CN106533966A - Network service resource arranging method and apparatus - Google Patents

Abstract

The invention provides a network service resource arrangement method and device, and relates to the field of communication networks. Wherein, the method includes: formally describing the network topology and the network service respectively, and obtaining a first model corresponding to the network topology and a second model corresponding to the network service; according to the first model and the obtained The second model respectively constrains the network traffic path and the network service deployment node; determines the optimization goal of the resource arrangement of the network service according to the constraints of the network traffic path and the network service deployment node; adopts a greedy algorithm according to the optimization goal Optimizing is performed to obtain the resource arrangement result of the network service. The invention can not only improve the resource utilization rate of the network service node, but also realize the load balance of the network link flow, and support the dynamic expansion of the network service resource.

Description

Resource orchestration method and device for network service

technical field

The present invention relates to the technical field of network communication, in particular, to a method and device for resource arrangement of network services.

Background technique

The function of computer network is divided into two main aspects: network connection and network service. Network connectivity refers to routing to connect different terminal devices. In the face of network path failure, dynamic routing protocols need to be implemented to ensure fault tolerance. In the face of network link bandwidth limitations, traffic engineering adjustments are required to avoid link congestion. The network service provides terminal devices with functional services of the fourth to seventh layers of the network, and the intermediate device carrying the network service is deployed between the terminal devices to process the passing user traffic. Network services can provide security detection, traffic statistics, network acceleration and other functions for terminal equipment. Web services are usually configured as functional sequences, providing a combination of multiple web services in a given order.

Software-defined networking and network function virtualization are important technologies in the current network development. Software-defined networking implements fine-grained control of network traffic through open interfaces and logically centralized control of forwarding devices, and deploys corresponding rules in forwarding devices. Network function virtualization refers to the use of virtualization technology to run the network function of proprietary intermediate equipment on a general-purpose processing platform to realize the dynamic allocation and expansion of network function resources.

Existing web services adopt static resource orchestration method. Restricted by the closure of hardware devices, network services are deployed at gateways or network traffic convergence points. By pre-evaluating the processing capabilities of intermediate devices required at deployment points, intermediate devices with corresponding processing performance are enabled. In addition, existing network services can achieve load balancing of intermediate devices with the help of software-defined network technology. For the statically enabled intermediate devices, the network traffic path can be selected through traffic traction, so that different intermediate devices can process the network traffic between different users, and the network traffic can be evenly distributed on the corresponding intermediate devices according to the order of consumption of processing power. Realize load balancing of intermediate devices.

Due to technical limitations, the existing resource orchestration methods for network services cannot achieve multi-objective optimization, and specifically have the following functional defects:

(1) The static resource orchestration method is deployed at the gateway node, and the network traffic is large, which requires the support of intermediate equipment with high processing performance, and the cost is relatively high.

(2) The static resource orchestration method, when the network traffic continues to grow and the existing intermediate equipment cannot provide sufficient processing capacity, it needs to be vertically expanded and replaced with an intermediate equipment with higher processing performance, which has poor scalability and high upgrade costs.

(3) The static resource orchestration method uses traffic traction to increase the consumption of link bandwidth, but does not constrain the resource consumption of link bandwidth, and there is a problem of link congestion.

Contents of the invention

The purpose of the present invention is to provide a resource orchestration method and device for network services. Wherein, the technical problem to be solved by the method is: how to reduce the multi-point failure probability of network services and avoid network link congestion.

In order to achieve the above purpose, the present invention provides a resource orchestration method for network services. The methods include:

Formally describing the network topology and the network service respectively, to obtain a first model corresponding to the network topology and a second model corresponding to the network service;

Constraining network traffic paths and network service deployment nodes respectively according to the first model and the second model;

determining an optimization target of resource orchestration of the network service according to the network traffic path and the constraints of the network service deployment node;

The greedy algorithm is used for optimization according to the optimization objective, and the resource arrangement result of the network service is obtained.

Optionally, the first model is:

G=(V,E,C)

Wherein, G represents the network topology, V represents a collection of network nodes, E represents a collection of network links, and C represents a collection of terminal traffic bandwidth.

Optionally, the second model is:

Among them, D _st represents the network service corresponding to the traffic T _st from terminal s to terminal t, Indicates the deployment order of the network function mi in the function sequence, N represents an integer greater than or equal to zero, s, _t∈H , H represents the set of end hosts, and n represents an integer greater than or equal to 1.

Optionally, the constraints of the network traffic path are:

Among them, s, t ∈ H, H represents the set of terminal hosts, i, j ∈ V, V represents the set of network nodes, (i, j) ∈ E, E represents the set of network links, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (i, j), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (i, j) j), k∈V, (j,k)∈E, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (j, k), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (j, k) k), is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through network node i, taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through network node i, and max(*, *) means The larger of two values, (j,i)∈E, It is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (j, i), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (j, i) i), C _st represents the network link bandwidth resources occupied by the traffic T _st from terminal s to terminal t, and L represents the upper bound of network link bandwidth resources used by network link (i, j).

Optionally, the constraints of the network service deployment node are:

Among them, s, t ∈ H, H represents the set of terminal hosts, i, j ∈ V, V represents the set of network nodes, m _p ∈ W, W represents the set of network functions, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through network node i, taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through network node i, is a 0-1 variable, taking a value of 0 means that network node i does not use the network function m _p to process the traffic T _st from terminal s to terminal t, and taking a value of 1 means that network node i uses the network function m _p to process traffic from terminal s to terminal t The flow T _st of terminal t, Denotes the deployment order of the network function m _p in the function sequence, and represent the auxiliary variables, respectively, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (i, j), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (i, j) j), N represents an integer greater than or equal to zero, m _q ∈ W, Indicates the deployment order of the network function m _q in the function sequence, is a 0-1 variable, taking a value of 0 means that network node i does not use the network function m _q to process the traffic T _st from terminal s to terminal t, and taking a value of 1 means that network node i uses the network function m _q to process traffic from terminal s to terminal t The flow T _st of terminal t, Indicates the computing resources consumed by the network function m _p to process the traffic T _st from the terminal s to the terminal t, B _i represents the total computing resources at the network node i, M represents the maximum value that can be expressed in the computer, and K represents the deployed network The total number of service nodes, z _i is a 0-1 variable, a value of 0 indicates that no network service is deployed at network node i, and a value of 1 indicates that network service is deployed at network node i.

Optionally, the optimization objective is:

min(λ ₁ L+λ ₂ K)

Among them, min(*) means seeking the minimum value, L means the upper bound of network link bandwidth resources used by network link (i, j), K means the total number of nodes deployed with network services, λ ₁ and λ ₂ are administrators Or a parameter set by an application program, and λ ₁ +λ ₂ =1.

Optionally, the method also includes:

When the network service rule of the network service changes or the processing capability required by the network service increases, the network service deployment node is adjusted, and the network traffic path is incrementally updated.

Correspondingly, the present invention also provides a network service resource arrangement device. The devices include:

A description unit, configured to formally describe the network topology and the network service respectively, to obtain a first model corresponding to the network topology and a second model corresponding to the network service;

a constraint unit, configured to respectively constrain network traffic paths and network service deployment nodes according to the first model and the second model;

A determining unit, configured to determine an optimization target of resource orchestration of the network service according to the network traffic path and the constraints of the network service deployment node;

The optimization unit is configured to perform optimization by using a greedy algorithm according to the optimization target, and obtain the resource arrangement result of the network service.

Optionally, the first model is:

G=(V,E,C)

Wherein, G represents the network topology, V represents a collection of network nodes, E represents a collection of network links, and C represents a collection of terminal traffic bandwidth.

Optionally, the second model is:

Among them, D _st represents the network service corresponding to the traffic T _st from terminal s to terminal t, Indicates the deployment order of the network function mi in the function sequence, N represents an integer greater than or equal to zero, s, _t∈H , H represents the set of end hosts, and n represents an integer greater than or equal to 1.

It can be seen from the above technical solution that the network topology and the network service are formally described respectively to obtain a first model corresponding to the network topology and a second model corresponding to the network service; and according to the first model and the obtained The second model respectively constrains the network traffic path and the network service deployment node; and then determines the optimization target of resource orchestration of the network service according to the constraints of the network traffic path and the network service deployment node; and then adopts according to the optimization target The greedy algorithm is optimized to obtain the resource orchestration result of the network service, which not only plans the network traffic path, but also determines the network service deployment node, which can achieve high utilization of computing resources and load balancing of link bandwidth resources, thereby reducing network traffic. Multi-point failure probability of services and avoid congestion of network links.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the description of the embodiments or the prior art. Apparently, the drawings in the following description are only some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

FIG. 1 is a flow chart of a resource orchestration method for network services provided by an embodiment of the present invention;

Fig. 2 is a schematic diagram of a network topology provided by an embodiment of the present invention;

Fig. 3 is a schematic diagram of a network service rule provided by an embodiment of the present invention;

FIG. 4 is a schematic structural diagram of a resource orchestration device for network services provided by an embodiment of the present invention;

Fig. 5 is a schematic structural diagram of a resource orchestration device for network services provided by another embodiment of the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

FIG. 1 is a flow chart of a resource orchestration method for network services provided by an embodiment of the present invention. As shown in FIG. 1, a method for orchestrating resources of a network service provided by an embodiment of the present invention includes:

In step S101, the network topology and the network service are formally described respectively to obtain a first model corresponding to the network topology and a second model corresponding to the network service.

Wherein, the first model is:

G=(V,E,C)

Wherein, G represents the network topology, V represents a collection of network nodes, E represents a collection of network links, and C represents a collection of terminal traffic bandwidth.

Wherein, the second model is:

Among them, D _st represents the network service corresponding to the traffic T _st from terminal s to terminal t, Indicates the deployment order of the network function mi in the function sequence, N represents an integer greater than or equal to zero, s, _t∈H , H represents the set of end hosts, and n represents an integer greater than or equal to 1.

Specifically, the network topology includes host distribution, forwarding device distribution, and network link bandwidth. Formally describe the network topology as G=(V,E,C), V is a collection of network nodes, E is a collection of network links, Represents the collection of terminal hosts, and C is the collection of terminal traffic bandwidth. Network services involve various network functions such as firewall, intrusion detection, traffic statistics and application proxy, and different network functions are combined in a specific sequence to provide users with network services. Formally describe the network service, the network function set W={m ₁ ,m ₂ ,…,m _n } can be obtained, the traffic T _st (s,t∈H) corresponding to the network service from terminal s to terminal t is Represents the deployment order of the network function _mi in the function sequence, The order of values is hour first, A value of 0 indicates that T _st is not processed by network function _mi .

Next, in step S102, network traffic paths and network service deployment nodes are constrained respectively according to the first model and the second model.

Wherein, the constraints of the network traffic path are:

Among them, s, t ∈ H, H represents the set of terminal hosts, i, j ∈ V, V represents the set of network nodes, (i, j) ∈ E, E represents the set of network links, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (i, j), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (i, j) j), k∈V, (j,k)∈E, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (j, k), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (j, k) k), is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through network node i, taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through network node i, and max(*, *) means The larger of two values, (j,i)∈E, It is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (j, i), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (j, i) i), C _st represents the network link bandwidth resources occupied by the traffic T _st from terminal s to terminal t, and L represents the upper bound of network link bandwidth resources used by network link (i, j).

Wherein, the constraints of the network service deployment node are:

Among them, s, t ∈ H, H represents the set of terminal hosts, i, j ∈ V, V represents the set of network nodes, m _p ∈ W, W represents the set of network functions, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through network node i, taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through network node i, is a 0-1 variable, taking a value of 0 means that network node i does not use the network function m _p to process the traffic T _st from terminal s to terminal t, and taking a value of 1 means that network node i uses the network function m _p to process traffic from terminal s to terminal t The flow T _st of terminal t, Denotes the deployment order of the network function m _p in the function sequence, and represent the auxiliary variables, respectively, is a 0-1 variable, taking a value of 0 means that the traffic T _st from terminal s to terminal t does not pass through the network link (i, j), and taking a value of 1 means that the traffic T _st from terminal s to terminal t passes through the network link (i, j) j), N represents an integer greater than or equal to zero, m _q ∈ W, Indicates the deployment order of the network function m _q in the function sequence, is a 0-1 variable, taking a value of 0 means that network node i does not use the network function m _q to process the traffic T _st from terminal s to terminal t, and taking a value of 1 means that network node i uses the network function m _q to process traffic from terminal s to terminal t The flow T _st of terminal t, Indicates the computing resources consumed by the network function m _p to process the traffic T _st from the terminal s to the terminal t, B _i represents the total computing resources at the network node i, M represents the maximum value that can be expressed in the computer, and K represents the deployed network The total number of service nodes, z _i is a 0-1 variable, a value of 0 indicates that no network service is deployed at network node i, and a value of 1 indicates that network service is deployed at network node i.

Specifically, for the constraints on the network traffic path, it is set that the traffic between terminals is single-path and acyclic. Aiming at the constraints of network service deployment nodes, network services satisfy order preservation and uniqueness. In addition, by introducing auxiliary variables, the network service deployment node constraints are modeled. The order of the auxiliary variables increases with the network traffic from the source node to the destination node. The function parameters of the service rules are in the same order as the auxiliary variables.

Next, in step S103, an optimization target of resource orchestration of the network service is determined according to the network traffic path and constraints of the network service deployment node.

Wherein, the optimization objective is:

min(λ ₁ L+λ ₂ K)

Among them, min(*) means seeking the minimum value, L means the upper bound of network link bandwidth resources used by network link (i, j), K means the total number of nodes deployed with network services, λ ₁ and λ ₂ are administrators Or a parameter set by an application program, and λ ₁ +λ ₂ =1.

Specifically, the resource orchestration of network services needs to meet various optimization objectives. In order to achieve load balancing of network link traffic, there is L _min , that is, to minimize the maximum load of network links. In order to improve the resource utilization of network service nodes, avoid Decentralized node deployment increases the failure rate, and there is K _min , which minimizes the total number of network service deployment nodes.

Transform the multi-objective optimization problem into a single-objective optimization problem, and the optimization objective is:

min(λ ₁ L+λ ₂ K)

Among them, λ ₁ and λ ₂ are the parameters set by the administrator or the application program. If not manually set, the default λ ₁ =K _min /(L _min +K _min ), λ ₂ =L _min /(L _min +K min _min ), L _min and K _min are the minimum values taken in the single-objective optimization problem.

Finally, in step S104, a greedy algorithm is used for optimization according to the optimization target, and a resource arrangement result of the network service is obtained.

Specifically, the optimization process using the greedy algorithm is as follows:

S1. Initialize an unorganized network service rule set, including all input rules, and each rule specifies a logical network service sequence required by traffic between certain two terminals. Initializes the set of orchestrated web service rules, which is an empty set.

S2. If the unorganized network service rule set is empty, go to step S12; otherwise, select the rule r that requires the least network service processing capacity in the unorganized network service rule set.

S3. Select the network service m in the rule r in sequence.

S4. Select the node with the most remaining computing resources among the nodes that have deployed the network service, and deploy the network service m. If it is feasible, go to S7. Otherwise, the remaining resources of the existing deployed nodes do not meet the deployment conditions.

S5. Select the node with the most computing resources among the nodes that have not deployed the network service, deploy the network service m, if feasible, go to S7, otherwise the rule r cannot be deployed.

S6. Mark the rule r as undeployable and put it into the orchestrated set, and go to S2.

S7. Steps S3-S6 are repeated until all deployment nodes of the network service in rule r are determined.

S8. Form the deployment nodes of the network service in the rule r into a sequence P, and add the source terminal s and the destination terminal t in the rule r to the beginning and end of the sequence P respectively.

S9. Select a network node pair (p _i , p _i+1 ) i=1, 2, . . . , |P|-1 in the sequence P sequentially.

S10. Select the path with the lowest bandwidth resource used by the network link among all the paths from the network node p _i to the node p _i+1 , and deploy the corresponding forwarding rule.

S11. Repeat steps S9-S10 until all network node pairs in the sequence P have reachable paths, and go to step S2.

S12. The network service deployment nodes and corresponding network traffic paths of all input rules are determined, and the network service resource arrangement is completed.

Preferably, the method further includes: when the network service rule of the network service changes or the processing capability required by the network service increases, adjusting the network service deployment node and incrementally updating the network traffic path. In this way, dynamic expansion of network services is supported

Specifically, the resource orchestration method for network services provided in this embodiment can adapt to changes in resource constraints and service rules, and dynamically update deployment results. In terms of resource constraints, when the network service processing capacity consumed by the deployment rules increases and exceeds the upper limit of the processing capacity of the current deployment node, a new deployment node can be selected according to steps S4-S5, and the updated network node can be selected according to step S10 and adjacent deployments in the sequence Forwarding paths between nodes. In terms of service rules, there are three cases of modifying, adding, and deleting service rules. For deleting a service rule, it means deleting the corresponding deployed forwarding rules and network service resources on the deployment node. For adding a service rule, follow steps S3-S11. For Modifying the rules is done in two steps: deleting the original service rules and adding new and modified service rules.

In this embodiment, a first model corresponding to the network topology and a second model corresponding to the network service are obtained by formally describing the network topology and the network service respectively; and according to the first model and the second model The two models respectively constrain the network traffic path and the network service deployment node; then determine the optimization goal of the resource arrangement of the network service according to the constraints of the network traffic path and the network service deployment node; and then adopt a greedy algorithm according to the optimization goal Optimizing to obtain the resource orchestration result of the network service, not only planning the network traffic path, but also determining the network service deployment node, can achieve high utilization of computing resources and load balancing of link bandwidth resources, thereby reducing network service Multi-point failure probability, and avoid congestion of network links.

Fig. 2 is a schematic diagram of a network topology provided by an embodiment of the present invention. Fig. 3 is a schematic diagram of a network service rule provided by an embodiment of the present invention. The specific embodiment of the present invention will be further described in detail with reference to FIG. 2 and FIG. 3 . The network topology in Figure 2 and the network service rules in Figure 3 can be applied to cloud data centers, operator networks, and enterprise networks. Figure 2 contains 4 terminals H1-H4 and 6 forwarding devices SW1-SW6, where the upper limit of the processing capability of network services deployed at SW5 and SW6 is 10, and the upper limit of computing capabilities of network services deployed at SW1-SW4 is 8. As shown in Figure 3, security detection, firewall FW and intrusion detection IDS functions are provided for the traffic from H1 to H2. H3 traffic provides security detection and network acceleration services, FW, IDS and proxy function Proxy, the traffic size of H1 to H3 is 2, the processing capacity of FW and IDS is required to be 4 and 6, and the processing capacity of Proxy is 3; for H1 to H4 Traffic provides network acceleration services, proxy, the traffic size of H1 to H4 is 4, and the proxy processing capacity is 6.

As a result, the network topology is:

G=(V,E,C)

V={H1,H2,H3,H4,S1,S2,S3,S4,S5,S6}

H={H1,H2,H3,H4}

E={(H1,S1),(S1,S2),(S1,S5),(S5,S6),(S6,S2),(S6,S4),(S3,S4),(S3,S5) ,(H2,S2),(H3,S3),(H4,S4)},

C={C _H1H2 =1, C _H1H3 =2, C _H1H4 =4}

Among them, C _H1H2 represents the network link bandwidth resource occupied by the flow from terminal _H1 to terminal _H2 , C _H1H3 represents the network link bandwidth resource occupied by the flow from terminal _H1 to terminal _H3 , and C _H1H4 represents the terminal _H1 Network link bandwidth resource occupied by traffic to terminal H4 _.

As a result, the network service is:

W={m ₁ =FW, m ₂ =IDS, m ₃ =Proxy}

web service rules

web service rules

web service rules

Among them, D _st represents the network service corresponding to the traffic T _st from terminal s to terminal t, Denotes the deployment order of network function _mi in the function sequence.

As a result, the constraints of the network traffic path are:

From this, the constraints of network service deployment nodes are:

As a result, the optimization objective is:

min0.2L+0.8K

As a result, the optimization process using the greedy algorithm is specifically:

L1. Initialize the unorganized network service rule set Φ = {D _H1H2 , D _H1H3 , D _H1H4 }, the programmed network service rule set

L2. If Φ is empty, go to step L12, otherwise select the rule r in Φ that requires the least network service processing capacity, that is, first select D _H1H2 .

L3. Select the network service m in the rule r in sequence, and for D _H1H2 , select the FW and IDS in sequence.

L4. Select the network node that has the most remaining computing resources among the nodes that have already deployed network services, and deploy network service m. If it is feasible, go to L7. Otherwise, the remaining resources of the existing deployed nodes do not meet the deployment conditions. For D _H1H2 The FW has no existing deployment node to proceed to the next step. The remaining resource of the existing deployment node SW5 of the IDS in D _H1H2 is at most 8 (the FW of D _H1H2 has already been deployed), and the IDS can be deployed.

L5. Select the node with the most resources among the nodes that have not deployed the network service, deploy the network service m, if feasible, go to L7, otherwise the rule r cannot be deployed, and select the node SW5 for the FW in D _H1H2 .

L6. Mark the rule r as undeployable and put it into the orchestrated set, and turn to L2.

L7. Repeat steps L3-L6 until all deployment nodes of the network service in rule r are determined.

L8. The deployment nodes of the network service in the rule r form a sequence P, and the source terminal s and the destination terminal t in the rule r are respectively added to the beginning and end of the sequence P. For D _H1H2 ,

L9. Select the node pair (p _i , p _i+1 )i=1,2,...,|P|-1 in the sequence P in sequence, and pair The node pair is (H1,SW5 _FW )(SW5 _FW ,W5 _IDS )(W5 _IDS ,H2).

L10. Select the path with the lowest bandwidth resource used by the network link among all the paths from the network node p _i to the node p _i+1 , and deploy the corresponding forwarding rule. The path of (H1, SW5 _FW ) is H1-SW1-SW5, (SW5 _FW , SW5 _IDS ) path is SW5.

L11. Repeat steps L9-L10 until there are reachable paths between all node pairs in the sequence P, and go to L2.

L12. The network service deployment nodes and corresponding network traffic paths of all input rules are determined, and the resource arrangement of network services is completed.

The resource orchestration method for network services provided in this embodiment can adapt to changes in resource constraints and service rules, and dynamically update deployment results. In terms of resource constraints, when the network service processing capacity consumed by the deployed rules increases and exceeds the upper limit of the processing capacity of the current deployment node, a new deployment node can be selected according to steps L4-L5, and the updated network node and adjacent deployment in the sequence can be selected according to step L10 Forwarding paths between nodes. In terms of service rules, there are three cases of modifying, adding, and deleting service rules. For deleting a service rule, it means deleting the corresponding deployed forwarding rules and network service resources on the deployment node. For adding a service rule, follow steps L3-L11. For Modifying the rules is done in two steps: deleting the original service rules and adding new and modified service rules.

The embodiment of the present invention is based on software-defined network and network function virtualization technology. In order to enable the computer network to provide users with various network services, by jointly modeling the network topology and network service sequence, the network link bandwidth resources and supporting network services The deployed computing resources are orchestrated, network traffic paths are planned, and network service deployment nodes are determined to achieve high utilization of computing resources and load balancing of link bandwidth resources.

In a specific implementation manner, the embodiments of the present invention can be applied to operator networks, data center networks, and enterprise networks to provide users with stable and scalable network services, optimize network link loads, and save network operating costs.

For the method embodiment, for the sake of simple description, it is expressed as a series of action combinations, but those skilled in the art should know that the embodiment of the present invention is not limited by the described action order, because according to the embodiment of the present invention , certain steps may be performed in other order or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification belong to preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

Fig. 4 is a schematic structural diagram of a resource orchestration device for network services provided by an embodiment of the present invention. As shown in FIG. 4, the resource orchestration device for network services provided by an embodiment of the present invention includes a description unit 201, a constraint unit 202, a determination unit 203, and an optimization unit 204, wherein:

A description unit 201, configured to formally describe the network topology and the network service respectively, to obtain a first model corresponding to the network topology and a second model corresponding to the network service;

A constraining unit 202, configured to constrain network traffic paths and network service deployment nodes respectively according to the first model and the second model;

A determining unit 203, configured to determine an optimization target of resource orchestration of the network service according to the network traffic path and the constraints of the network service deployment node;

The optimization unit 204 is configured to use a greedy algorithm to optimize according to the optimization target, and obtain a resource arrangement result of the network service.

The network service resource orchestration device provided in this embodiment is applicable to the network service resource orchestration method corresponding to the above embodiments, and will not be repeated here.

This embodiment provides a resource orchestration device for network services. The description unit 201 is used to formally describe the network topology and the network service respectively, and obtain the first model corresponding to the network topology and the first model corresponding to the network service. Two models; the constraining unit 202 is configured to constrain the network traffic path and the network service deployment node respectively according to the first model and the second model; the determining unit 203 is configured to constrain the network traffic path and the network service deployment node according to the network traffic path The constraint determines the optimization goal of the resource arrangement of the network service; the optimization unit 204 is used to optimize according to the optimization goal using a greedy algorithm to obtain the resource arrangement result of the network service, which not only plans the network traffic path, but also determines the network traffic path. Service deployment nodes can achieve high utilization of computing resources and load balancing of link bandwidth resources, thereby reducing the probability of multi-point failure of network services and avoiding network link congestion.

In an optional embodiment of the present invention, the first model is:

G=(V,E,C)

Wherein, G represents the network topology, V represents a collection of network nodes, E represents a collection of network links, and C represents a collection of terminal traffic bandwidth.

In an optional embodiment of the present invention, the second model is:

Among them, D _st represents the network service corresponding to the traffic T _st from terminal s to terminal t, Indicates the deployment order of the network function mi in the function sequence, N represents an integer greater than or equal to zero, s, _t∈H , H represents the set of end hosts, and n represents an integer greater than or equal to 1.

Fig. 5 is a schematic structural diagram of a resource orchestration device for network services provided by another embodiment of the present invention. As shown in FIG. 5 , the resource orchestration device of the network service includes: a processor (processor) 301, a memory (memory) 302 and a communication bus 304;

Wherein, the processor 301 and the memory 302 complete mutual communication through the communication bus 304;

The processor 301 is used to call the program instructions in the memory 302 to execute the methods provided by the above method embodiments, for example, including: formally describing the network topology and network services respectively, and obtaining The corresponding first model and the second model corresponding to the network service; according to the first model and the second model respectively constrain the network traffic path and the network service deployment node; according to the network traffic path and the network The constraint of the service deployment node determines the optimization target of the resource arrangement of the network service; according to the optimization target, a greedy algorithm is used for optimization to obtain the resource arrangement result of the network service.

This embodiment discloses a computer program product, the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by the computer, the computer The method provided by each of the foregoing method embodiments can be executed, for example, including: formally describing the network topology and the network service respectively, and obtaining a first model corresponding to the network topology and a second model corresponding to the network service; Constraining network traffic paths and network service deployment nodes respectively according to the first model and the second model; determining an optimization target of resource orchestration of the network service according to constraints of the network traffic paths and the network service deployment nodes; The greedy algorithm is used for optimization according to the optimization objective, and the resource arrangement result of the network service is obtained.

This embodiment provides a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the methods provided in the above method embodiments, for example, including : formally describe the network topology and the network service respectively, obtain a first model corresponding to the network topology and a second model corresponding to the network service; constrain respectively according to the first model and the second model The network traffic path and the network service deployment node; determine the optimization target of resource orchestration of the network service according to the constraints of the network traffic path and the network service deployment node; optimize according to the optimization target using a greedy algorithm, and obtain the Resource orchestration results for web services.

Those of ordinary skill in the art can understand that all or part of the steps for realizing the above-mentioned method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the It includes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

The base station and other embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, Located in one place, or can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative effort.

Through the above description of the implementations, those skilled in the art can clearly understand that each implementation can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic discs, optical discs, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, rather than to limit them; although the embodiments of the present invention have been described in detail with reference to the foregoing embodiments, those skilled in the art The skilled person should understand that: it is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the present invention The scope of the technical solution of each embodiment of the embodiment.

Claims (10)

1. the resource method of combination of a kind of network service, it is characterised in that methods described includes：

Formalized description is carried out respectively to network topology and network service, obtain the first model corresponding with the network topology and The second model corresponding with the network service；

According to first model and second model difference constraint network flow path and Web Service Deployment node；

Determine that according to the constraint of the network traffics path and the Web Service Deployment node resource of the network service is compiled The optimization aim of row；

It is optimized using greedy algorithm according to the optimization aim, obtains the resource layout result of the network service.

2. the resource method of combination of network service according to claim 1, it is characterised in that first model is：

G=(V, E, C)

Wherein, G represents the network topology, and V represents the set of network node, and E represents the set of network link, and C represents terminal The set of flow bandwidth.

3. the resource method of combination of network service according to claim 1, it is characterised in that second model is：

$D_{st}=\left\{d_i^{st}\right|d_i^{st}\∈N,0\≤d_i^{st}\≤n,i=1,...,n\}$

Wherein, D_stRepresent terminal s to the flow T of terminal t_stCorresponding network service,Represent network function m_iIn functional sequence In deployment order, N represents the integer more than or equal to zero, s, t ∈ H, and H represents the set of end host, and n is represented and is more than or waits In 1 integer.

4. the resource method of combination of network service according to claim 1, it is characterised in that the network traffics path It is constrained to：

$\∀s,t\∈H:\underset{i\∈V:(i,j)\∈E}{\Σ}{x}_{ij}^{st}-\underset{i\∈V:(j,k)\∈E}{\Σ}{x}_{jk}^{st}=\left\{\begin{array}{c}1,j=t\\ 0,j\≠s,t,j\∈V\\ -1,j=s\end{array}\right.$

$\∀s,t\∈Handi\∈V:y_i^{st}=max(\underset{j\∈V:(j,i)\∈E}{\Σ}{x}_{ji}^{st},\underset{j\∈V:(i,j)\∈E}{\Σ}{x}_{ij}^{st})$

$\∀s,t\∈Handi\∈V:y_i^{st}\≤1$

$\∀i,j\∈Vand(i,j)\∈E:\underset{s,t\∈H}{\Σ}{x}_{ij}^{st}.C_{st}\≤L$

Wherein, s, t ∈ H, H represent the set of end host, i, j ∈ V, and V represents the set of network node, and (i, j) ∈ E, E are represented The set of network link,For 0-1 variables, take 0 value and represent terminal s to the flow T of terminal t_stWithout network link (i, j), Taking 1 value represents terminal s to the flow T of terminal t_stThrough network link (i, j), k ∈ V, (j, k) ∈ E,For 0-1 variables, 0 is taken Value represents terminal s to the flow T of terminal t_stWithout network link (j, k), take 1 value and represent terminal s to the flow T of terminal t_st Through network link (j, k),For 0-1 variables, take 0 value and represent terminal s to the flow T of terminal t_stWithout network node i, Taking 1 value represents terminal s to the flow T of terminal t_stThrough network node i, max (*, *) represents the higher value among two values, (j, i) ∈ E,For 0-1 variables, take 0 value and represent terminal s to the flow T of terminal t_stWithout network link (j, i), 1 value is taken Represent terminal s to the flow T of terminal t_stThrough network link (j, i), C_stRepresent terminal s to the flow T of terminal t_stShared Network link bandwidth resources, L represent the upper bound of the network link (i, j) using network link bandwidth resources.

5. the resource method of combination of network service according to claim 1, it is characterised in that the Web Service Deployment section That what is put is constrained to：

$\∀s,t\∈Handi\∈Vand{m}_p\∈W:g_{ip}^{st}\≤y_i^{st}$

$\∀s,t\∈Handi\∈Vand{m}_p\∈Wand{d}_p^{st}\≠0:\underset{i\∈V}{\Σ}{g}_{ip}^{st}=1$

$\∀s,t\∈Handi,j\∈V:u_i^{st}-u_j^{st}+\left|V\right|x_{ij}^{st}\≤\left|V\right|-1$

$\∀s,t\∈Handi\∈V:1\≤u_i^{st}\≤\left|V\right|,u_i^{st}\∈N$

$\∀s,t\∈Handi,j\∈Vand{m}_p,m_q\∈W:(d_q^{st}-d_p^{st})(u_i^{st}-u_j^{st})\≤\left|W\right|\left|V\right|(2-g_{ip}^{st}-g_{iq}^{st})$

$\∀i\∈V:\underset{p\∈N}{\Σ}\underset{s,t\∈H}{\Σ}{g}_{ip}^{st}{A}_p^{st}\≤B_i$

$\∀i\∈V:\underset{p\∈N}{\Σ}\underset{s,t\∈H}{\Σ}{g}_{ip}^{st}\≤{\mathrm{Mz}}_i$

$\underset{i\∈V}{\Σ}{z}_i=K$

Wherein, s, t ∈ H, H represent the set of end host, and i, j ∈ V, V represent the set of network node, m_p∈ W, W represent network The set of function,For 0-1 variables, take 0 value and represent terminal s to the flow T of terminal t_stWithout network node i, 1 value table is taken Show terminal s to the flow T of terminal t_stThrough network node i,For 0-1 variables, take 0 value and represent unused net at network node i Network function m_pProcess the flow T from terminal s to terminal t_st, take 1 value and represent at network node i using network function m_pProcess from end Flow Ts of the end s to terminal t_st,Represent network function m_pDeployment order in functional sequence,WithAuxiliary is represented respectively Variable,For 0-1 variables, take 0 value and represent terminal s to the flow T of terminal t_stWithout network link (i, j), take 1 value and represent Flow T of terminal s to terminal t_stThrough network link (i, j), N represents the integer more than or equal to zero, m_q∈ W,Represent net Network function m_qDeployment order in functional sequence,For 0-1 variables, take 0 value and represent uses no network function at network node i m_qProcess the flow T from terminal s to terminal t_st, take 1 value and represent at network node i using network function m_qProcess from terminal s to The flow T of terminal t_st,Represent network function m_pProcess the flow T from terminal s to terminal t_stThe computing resource of required consumption, B_iComputing resource total at network node i is represented, the maximum that M can be represented in representing computer, K are represented and be deployed with network service Node sum, z_iFor 0-1 variables, take 0 value and represent and at network node i, do not dispose any network service, take 1 value and represent network Network service is deployed with node i.

6. the resource method of combination of network service according to claim 1, it is characterised in that the optimization aim is：

min(λ₁L+λ₂K)

Wherein, min (*) represents and minimizes that L represents the upper bound of the network link (i, j) using network link bandwidth resources, K tables Show the sum of the node for being deployed with network service, λ₁With λ₂For keeper or the parameter of application setting, and λ₁+λ₂=1.

7. the resource method of combination of network service according to claim 1, it is characterised in that methods described also includes：

In the case where the disposal ability needed for the network service rule change of the network service or the network service increases, Adjust the Web Service Deployment node, and network traffics path described in incremental update.

8. the resource layout device of a kind of network service, it is characterised in that described device includes：

Description unit, for carrying out formalized description respectively to network topology and network service, obtains and the network topology pair The first model answered and the second model corresponding with the network service；

Constraint element, for according to first model and second model difference constraint network flow path and network service Deployment node；

Determining unit, for determining the network according to the constraint in the network traffics path and the Web Service Deployment node The optimization aim of the resource layout of service；

Optimization unit, for being optimized using greedy algorithm according to the optimization aim, obtains the resource of the network service Layout result.

9. the resource layout device of network service according to claim 8, it is characterised in that first model is：

G=(V, E, C)

Wherein, G represents the network topology, and V represents the set of network node, and E represents the set of network link, and C represents terminal The set of flow bandwidth.

10. the resource layout device of network service according to claim 8, it is characterised in that second model is：

$D_{st}=\left\{d_i^{st}|d_i^{st}\∈N,0\≤d_i^{st}\≤n,i=1,...,n\right\}$

Wherein, D_stRepresent terminal s to the flow T of terminal t_stCorresponding network service,Represent network function m_iIn functional sequence In deployment order, N represents the integer more than or equal to zero, s, t ∈ H, and H represents the set of end host, and n is represented and is more than or waits In 1 integer.